Mechanisms of transfer of ammonia in smoke from ammonium compounds in tobacco
Yves SAINT JALM , G. DUVAL, T. CONTE and I. BONNICHON Altadis Research Center Fleury les Aubrais FRANCE Survey of the European market (1998)
• 53 leading brands in Europe – France (24), Netherlands, Belgium, Germany, Italy, Spain, Switzerland, Austria, Greece and UK – Exclusively Virginia and American blends – Full flavor (tar between 10 and 15 mg) • Survey of different chemical and physical parameters – Ammonium in tobacco, Ammonia in smoke Relationship between Ammonium in Tobacco and Ammonia in smoke
35,0
30,0
25,0
20,0 NH3 (µg/cig) NH3
15,0
10,0 0,00 0,05 0,10 0,15 0,20 0,25 0,30 0,35 0,40 NH3 (%) Relationship between Ammonium in Tobacco and Ammonia in smoke
• Linear regression – NH3 = 13.2 + 37.9 NH4 – NH3 in g/cig – NH4 in % in dry tobacco – r = 0.594; – ANOVA : p <0.01; significant relationship • Interpretation of results difficult because – differences in blend and cigarette design Transfer of Ammonia From Ammonium compounds
• These data seem to illustrate a direct transfer of Ammonia into smoke from Ammonium compounds in tobacco with a very low transfer rate • An increase of 0.2 % of Ammonium in blend • gives a mean increase of 8 g / cig of ammonia • from the regression equation • with a potential of 130 g / cig • assuming 800 mg of tobacco /cig • and a transfer rate of 8 % (equivalent to nicotine) Experimental design
• Hypotheses – Ammonium compounds produce Ammonia under thermal degradation – Ammonia reacts with carbonyl compounds produced by the pyrolysis of carbohydrates – with a special emphasis on sugars (Glucose, Fructose and Sucrose which degradation occurs at the same temperature range)
R O NH3 R N R N R H
R O R O R N R Experimental design Response surface type
• American blend – Addition of controlled amounts of DAP and
Sucrose DAP 0 % 1 % 2 % 0 % S1D1 S1D2 S1D3 Sucrose 2,5 % S2D1 S2D2 S2D3 5 % S3D1 S3D2 S3D3
– King size cigarette • Constant tobacco weight : 792 mg • no filter ventilation Controlled parameters Tobacco rod
• Total alkaloïds • Ammonium • Glucose, Fructose, Sucrose • Phosphate • Continuous Flow Analysis Tobacco - Ammonium
DAP0 DAP1 DAP2 m th m th m th
S0 0.30 0.30 0.52 0.53 0.72 0.76 S2.5 0.30 0.29 0.52 0.52 0.71 0.75 S5 0.29 0.29 0.49 0.51 0.71 0.73
% in dry tobacco Tobacco - Sucrose
DAP0 DAP1 DAP2 m th m th m th
S0 0.87 0.87 0.80 0.86 0.63 0.85 S2.5 2.67 3.29 2.53 3.26 2.67 3.22 S5 4.75 5.59 4.73 5.54 4.81 5.49
% in dry tobacco Tobacco - Glucose
DAP0 DAP1 DAP2 m th m th m th
S0 1.78 1.78 1.63 1.76 1.55 1.75 S2.5 1.80 1.74 1.73 1.72 1.77 1.70 S5 2.10 1.70 1.92 1.68 1.86 1.66
% in dry tobacco
Same pattern for Fructose Tobacco - Alcaloïds
DAP0 DAP1 DAP2 m th m th m th
S0 2.17 2.17 2.13 2.15 2.13 2.13 S2.5 2.12 2.12 2.06 2.10 2.05 2.08 S5 2.06 2.07 2.04 2.05 2.01 2.03
% in dry tobacco Tobacco analysis - Conclusions
• The experimental design was realised as planned • Sucrose was partially inverted to Glucose and Fructose during the process • Reaction between sugars and DAP may have occurred at this stage Transfer of Ammonia Variables
• Smoke – Tar, Nicotine • ISO methods – Pyrazines • GC – Diacetyl, Methylglyoxal • HPLC (derivatization with o plenylendiamine) – Ammonia • Ion exchange liquid Chromatography Methodology
• Variables expressed in ppm / NFDPM • Mathematical model – V = a + b x A + c x B + d x AB + e x A² + f x B² • where : A is added sucrose in % and B is added DAP in % • ANOVA – To simplify the model (exclude or not the second order terms in the model) • Calculate the final model and the estimated response surface Smoke - Ammonia
Estimated Response Surface
2900 2600 2300 R² = 0.841 2000 1700
Ammonia 1400 1.62 1100 0.81.2 0 1 0.4 DAP 2 3 4 0 Sucrose 5
NH3 = 1572.6 80 x S + 586.6 x DAP + 36.1 x S x DAP Smoke - MethylGlyoxal
Estimated Response Surface
3900 3500 3100 R² = 0.904 2700 2300 MeGlyoxal 1.62 1900 0.81.2 0 1 0.4 DAP 2 3 4 0 Sucrose 5
MeG = 2669.3 + 223 X S 370.1 x DAP 29.5 x S x DAP Smoke - Pyrazines
Estimated Response Surface
53 49 R² = 0.863 45 41
Pyrazines 37 1.62 33 0.81.2 0 1 0.4 DAP 2 3 4 0 Sucrose 5
Pyr = 33.5 + 0.78 x S + 6.2 x DAP + 0.2 x S x DAP Transfer of ammonia
• There is a direct transfer of ammonia in smoke from ammonium compounds • This transfer is partially inhibited when sucrose yields increase • Ammonia reacts with some carbonyl compounds produced by sugar pyrolysis Effect on Nicotine transfer Variables
• Nicotine • Expressed in % of DTPM • Nicotine in vapour phase • Expressed in % of total Nicotine • Denuder tube method • “Smoke pH” • Total smoke • Condensate – Aqueous solutions pH - Total smoke
DAP0 DAP2 DAP1
5,54 5,52 5,64
5,44 S0 5,51 5,51 5,53 S2.5 5,44 5,46 S5
No significant effect pH - TPM
DAP0 DAP2 DAP1
6,36 6,27 6,31
6,24 S0 6,30 6,23 6,26 S2.5 6,38 6,18 S5
No significant effect Nicotine % of Dry PM
Estimated Response Surface
8 7.8 R² = 0.824 7.6 7.4
Nicotine 7.2 1.62 7 0.81.2 0 1 0.4 DAP 2 3 4 0 Sucrose 5
Nic = 7.415 0.196 x S + 0.208 x DAP + 0.026 x S² Vapour Phase Nicotine % of total Nicotine
Estimated Response Surface
5.4 5.1 4.8 R² = 0.694 4.5 4.2 3.9 1.62 3.6 0.81.2 DAP 0 1 2 0.4 3 4 5 0
Vapour phase Nicotine Sucrose
VPNic = 4.92 0.048 x S 0.296 DAP 0.047 x S x DAP Effect on Nicotine transfer
• No modification of “smoke pH” – whatever the trapping method used • Very limited modification of Nicotine concentration in condensate • Combustibility effect ? • No increase of Nicotine in vapour phase • a decrease is observed instead Summary
• The direct transfer of Ammonia in smoke from Ammonium compounds is probably controlled by the ammonium yield and the sugars (and related compounds) yield • No significant effect on Nicotine transfer in smoke have been observed